Background

Ankylosing spondylitis (AS) is a chronic inflammatory rheumatic disease primarily affecting the sacroiliac joints and spine [1]. Inflammation of the spinal structures and progressive spinal changes in the vertebrae and surrounding tissue, is largely responsible for the decreased physical function and mobility experienced by patients with AS [2]. Studies of AS often describe functional disabilities and measures of disease activity, however, they less often report the quality of life experienced by patients with AS and how this is related to AS disease characteristics. Health-related quality of life (HRQoL) is a multi-dimensional concept including not only a person’s physical wellbeing, but also a person’s mental health and physical ability, both as an individual and as a participating member of the community. The Medical Outcome Survey Short Form-36 (SF-36) was designed for use in clinical practice and research, health policy evaluations, and general population surveys and is utilised in many different countries [3,4,5,6]. Yang et al. recently performed a meta-analysis based on of 38 studies assessing HRQoL using the SF-36 in patients with AS and found that they had significantly worse HRQoL compared to persons from general populations and that to measure HRQoL should be regarded as an essential part of the overall assessment of patients with AS [35].

The impact on HRQoL of living alone and/or to be unmarried has previously been investigated in other diseases and in the general population, and to the best of our knowledge, for the first time in this study of patients with AS. The literature displays inconsistent results. In patients with chronic diseases such as cancer and multiple sclerosis, living alone is mostly associated with worse HRQoL, which is in line with the findings in our study [36,37,38]. However, in a longitudinal study, decline in the MCS score in patients with diabetes mellitus was associated with not living alone [39]. In a study from China, participants living alone had worse HRQoL [40], while the opposite was reported from another part of China, Shanghai [41]. Women in the Nurses’ Health Study who lived alone had lower risk of decline in the MH and VT domains compared with those living with a spouse. Furthermore, contact with friends and relatives and level of social engagement significantly protected against a decline in MH in women living alone but not among women living with a partner [42]. Thus, the influence of living alone on HRQoL is complex and seems to be related to the persons’ state of health, culture and gender, among other factors.

The sex-stratified analyses in our study revealed some differences; the disease activity seemed to be a more important factor in the physical component of HRQoL in men while the civil state was more important for the mental component of HRQoL in women, which are aspects that can be taken into account in the management of the patients with AS.

There are some limitations to be acknowledged; first this study is cross-sectional and thus we cannot draw any conclusions about the variables identified as associated with worse HRQoL and causality. To investigate this a longitudinal study is required. Second, the SF-36 Swedish normative population database was created in the 1990s and HRQoL in the general population might have changed over the years. However, HRQoL in the general population assessed by the SF-36 was stable from 1996 to 2004 in Norway [43], our neighbouring country, indicating that the difference in time may not be a significant problem. Third, we chose to divide the patients with AS according to their median values of PCS and MCS. Since there is no established cutoff for better or worse HRQoL we used this pragmatic approach. Furthermore, the modest number of subjects in the sex-stratified analyses result in broad confidence intervals and thus results need to be interpreted with some caution. Strengths of this study comprise first, a well-characterised patient cohort with an adequate number of patients allowing subgroup analyses of the sexes to be carried out for the first time. Second, SF-36 scores in patients with AS were compared to those in a large sample of controls from the general population who were precisely matched on sex and age.

Possible implications of our findings include that we have identified variables associated with worse HRQoL, which are modifiable. Even though our study is cross-sectional, and thus not able to show causality, by using these variables as a guide, patients may be treated more efficiently, leading to reduction in disease activity, pain and fatigue, thus potentially improving both the PCS and MCS scores as has been shown in some clinical trials [44,45,46]. To live alone without a partner was also associated with worse HRQoL in particular in women in the MCS score. A focus on social activities and community support for women with AS might help to improve MCS scores in this subset of patients with AS.

Conclusions

In this study we show that the patients with AS had significantly lower HRQoL when assessed by the SF-36, compared with the general population. The PCS score was more affected than the MCS score in both sexes. Both demographic and disease-related factors were associated with worse HRQoL, with partial overlap for the PCS and MCS. There were some differences between sexes in the factors associated with HRQoL. By modifying factors such as ASDAS-CRP and fatigue, HRQoL may potentially be improved. There is a need for longitudinal studies investigating predictors related to the development of HRQoL in AS over time. We intend to investigate factors related to the change in HRQoL over a 5-year period in this same cohort of patients with AS in a future study.